Table 2. Performance assessment of the classification model using the CV procedure.
Dataset | 10-fold CV | C-index | Brier score | Log-rank P-value (geo.mean) |
---|---|---|---|---|
Training | 3-omics training (60%) | 0.73 ± 0.06 | 0.14 ± 0.01 | 1.40e−4 |
RNA only | 0.67 ± 0.07 | 0.15 ± 0.01 | 1.93e−3 | |
miRNA only | 0.65 ± 0.05 | 0.14 ± 0.01 | 5.84e−4 | |
Methylation only | 0.68 ± 0.09 | 0.15 ± 0.02 | 1.07e−3 | |
Test | 3-omics test (40%) | 0.64 ± 0.11 | 0.16 ± 0.02 | 2.92e−2 |
RNA only | 0.63 ± 0.16 | 0.16 ± 0.02 | 4.07e−2 | |
miRNA only | 0.62 ± 0.12 | 0.17 ± 0.02 | 3.96e−2 | |
Methylation only | 0.60 ± 0.14 | 0.18 ± 0.02 | 4.72e−2 |
CV, cross-validation